The tinyML Virtual Summit 2021 and the tinyML Research Symposium 2021 registration now open! View Program

Register Here

tinyML Summit 2021

Enabling ultra-low Power Machine Learning at the Edge

March 22-26, 2021 | Online

About

The tinyML Summit will be held virtually the week of March 22, 2021. We are in the process of re-envisioning our flagship event as a highly interactive online experience.

In conjunction with the Summit, we are also pleased to announce that we have added a new event for 2021: the tinyML Research Symposium.

The tinyML Summit is the premier annual gatherings of senior level technical experts and decision makers representing fast growing global tinyML community. This diverse ecosystem is composed of professionals from industry, academia, start-ups, and government labs worldwide working on leading-edge ultra-low power machine learning technologies for end-to-end (hardware –system –software applications full stack) solutions.

Program

Monday, March 22

Tutorials

All times Pacific Standard
  • 8:00 am – 8:15 am

    Open / Welcome
  • 8:15 am – 9:45 am

    Two tutorials in parallel

    Training a Magic Wand
    Pete Warden, Technical Lead, TensorFlow Lite Micro Open-source Project, Google

    This tutorial will show how to gather data, train, and deploy an IMU-based model for recognizing gestures on an Arduino Nano BLE Sense 33. It will use the Arduino IDE and Colab scripts to develop the model and will explain the feature generation needed to go from raw accelerometer and gyroscope data to input suitable for a neural network. Using TensorFlow Lite Micro and Arm's CMSIS-NN library, you will learn how to create a practical application from scratch. It is recommended that you purchase the Arduino TinyML Kit to be able to follow along virtually.

    Image sensors for low power applications
    Song Chen, Research Scientist, Facebook Reality Labs Research

    Image sensors are the front end of many computer-vision based input modalities. These human-machine input modalities usually need to run on a mobile platform which has stringent power requirement. This tutorial will cover both low power image sensor design from a designer’s perspective and some useful practices to save sensor power from a user’s perspective especially in ML applications. We will start by laying out basics including the operation principle of pixels, readout chain and other common blocks in an image sensor. Then, the trade-off between power consumption and general sensor performance will be discussed. Following the discussion, the effectiveness of power reduction techniques like subsampling, low frame rate, etc. and the impact on following ML processing stages will be evaluated with examples. Finally, an ultra-low power global-shutter digital pixel sensor developed at Facebook Reality Labs Research will be introduced.

  • 9:45 am – 10:00 am

    Break
  • 10:00 am – 11:30 pm

    Two tutorials in parallel

    Advanced network quantization and compression through the AI Model Efficiency Toolkit (AIMET)
    Abhijit Khobare, Director of Software Engineering at Qualcomm Technologies, Inc. (QTI) and Chirag Patel, Principal Engr./Mgr. in Corp. R&D AI Research team at Qualcomm Technologies, Inc. (QTI)

    AI is revolutionizing industries, products, and core capabilities by delivering dramatically enhanced experiences. However, the deep neural networks of today use too much memory, compute, and energy. To make AI truly ubiquitous, it needs to run on the end device within a tight power and thermal budget. Quantization and compression help address these issues. In this tutorial, we’ll discuss:

    • The existing quantization and compression challenges
    • Our research in novel quantization and compression techniques to overcome these challenges
    • How developers and researchers can implement these techniques through the AI Model Efficiency Toolkit

    Build Industrial-Grade tinyML applications with Edge Impulse!
    Jan Jongboom, Co-Founder CTO and Daniel Situnayake, Founding tinyML Engineer, Edge Impulse

    In this free live workshop, you will build a full tinyML application, end-to-end, using the latest best practices in embedded machine learning. You will learn how to collect a dataset, design and train a tiny (but accurate) model, evaluate its performance, optimize it for embedded use, and integrate it into a real embedded application running on a genuine MCU. The first 250 registrants will receive the hardware (Thunderboard Kit from Silicon Labs) for free, compliments of Edge Impulse.

    You will leave this workshop feeling confident that you can solve real world problems using state of the art tinyML—and have fun while you are doing it!

    Register for this Workshop on the Summit registration form (check the Workshop box). Once registration for the 250 kits is reached you may still register, you won’t get the free kit but you may purchase one at Silicon Labs Thunderboard Kit (pending availability).

Tuesday, March 23

Summit

All times Pacific Standard
  • 8:00 am - 8:15 am

    Welcome and Opening Remarks
  • 8:15 am – 9:00 am

    Keynote: Putting AI on a Diet: TinyML and Efficient Deep Learning

    Song Han, Assistant Professor, MIT EECS

  • 9:00 am – 9:45 am

    Keynote: Many shades of acceleration - an Open TinyML Platform Perspective

    Luca Benini, Chair of digital Circuits and systems, ETHZ and Full Professor at the University of Bologna

  • 9:45 am – 10:00 am

    Break
  • 10:00 am - 10:15 am

    Today's Breakout Pitches
  • 10:15 am – 11:00 am

    tiny Talks (consecutive)

    Compute-in-Memory Hardware Accelerator for Always-On TinyML
    Sameer Wadhwa, Senior Director, Qualcomm

    Supporting Tensorflow Lite MCU in tiny low power FPGAs
    Hoon Choi, Fellow, Lattice Semiconductor

  • 11:00 am – 12:00 pm

    Panel discussion: Opportunities at the Edge: Venture and tinyML

    Panel chairs: Kurt Keutzer, Full Professor, University of CA, Berkeley and Chris Rowen, VP of Engineering, Cisco

    Pushing machine learning into ultra-low-power applications at the edge isn’t just an academically compelling idea, it is a potentially disruptive shift in mass-market technology. In this panel we've gathered four distinguished venture capitalists to look at tinyML opportunities through an entrepreneurial lens. In particular we have asked them to consider:

    • What makes you interested in investment opportunities for machine learning at the edge?
    • What is your general advice to tech entrepreneurs: build a horizontal platform for broad application or target a particular vertical? 
    • What are some of the particular near-term opportunities for venture investment at the edge that you find especially exciting?
    • How might VC’s value a tinyML-startup – how much is it driven by the target market, the team, the technology, or the data?
    • And the $6.4B question: what is the future killer app at the edge?

  • 12:00 pm – 1:00 pm

    Breakout Sessions (consecutive)

    Session #1: Hardware Optimization

    Performing inference on BNNs with xcore.ai
    Laszlo Kindrat, Senior Technologist, XMOS

    Ultra-low Power and Scalable Compute-In-Memory AI Accelerator for Next Generation Edge Inference
    Behdad Youssefi, Chief Executive Officer, Areanna AI

    CUTIE: Multi-PetaOP/s/W Ternary DNN inference Engine for TinyML
    Moritz Scherer, PhD Student ETH Zürich

    Hardware aware Dynamic Inference Technology
    Urmish Thakker, Principal Engineer, SambaNova Systems, Hardware aware Dynamic Inference Technology

    Other Breakouts

    • tinyML applications
    • tinyML vision challenge
    • Birds of a feather #1
    • tinyML for Good -- Conservation
    Participants to be announced

Wednesday, March 24

Summit

All times Pacific Standard
  • 8:00 am - 8:15 am

    Opening and Award Announcements
  • 8:15 am – 9:00 am

    Keynote:  miliJoules for 1000 Inferences: Machine Learning Systems “on the Cheap”

    Diana Marculescu, University of Texas at Austin

  • 9:00 am – 9:45 am

    Keynote: Adaptive Neural Networks for Agile TinyML

    Sek Chai, Co-founder and CTO, Latent AI

  • 9:45 am – 10:00 am

    Break
  • 10:00 am – 10:15 pm

    Break out pitches
  • 10:15 am – 11:00 am

    tiny Talks (consecutive)

    Hardware aware Dynamic Inference Technology
    Vikrant Tomar, CTO, Fluent.ai

    Building Computer Vision Applications under Extreme Constraints: Lessons from the Field
    Koen Helwegen, Deep Learning Scientist, Plumerai

  • 11:00 am – 12:00 pm

    Panel discussion: tinyML inference SW – where do we go from here?

    Moderated by Ian Bratt, Distinguished Engineer, Arm and Ofer Dekel, Microsoft Partner and a Principal Research Manager, Microsoft Research AI

    Join a collection of industry experts as we discuss the current state and potential future of tinyML inference SW.  What is missing today, what new technologies will impact tinyML inference SW, and how do we go forward as a community?

  • 12:00 pm – 1:00 pm

    Breakout Sessions

    Session #2: Algorithms and Tools 

    Neutrino: A BlackBox Framework for Constrained Deep Learning Model OptimizationDavis Sawyer, CPO Deeplite Inc.,

    Hardware Aware Training for Efficient Keyword Spotting on General Purpose and Specialized HardwareChris Eliasmith, University of Waterloo,

    Low-precision Winograd Convolution over Residue Number SystemZhi-Gang Liu, Principal Research Engineer, Arm Research

    An introduction to an open-source fixed-point inference framework - NNoMJianjia Ma, Research Fellow, University of Southampton

    Other Breakouts

    • Sponsor Sessions #1
    • Birds of a feather #2
    • The tinyML Market
    • tinyML for Good -- STEM Education
    Participants to be announced

Thursday, March 25

Summit

All times Pacific Standard
  • 8:00 am - 8:15 am

    Awards handouts
  • 8:15 am – 9:00 am

    Keynote: Efficient Audio-Visual Understanding on AR Devices

    Vikras Chandra, On-device AI applied research and engineering organization lead, Facebook Reality Labs

  • 9:00 am – 9:45 am

    Keynotes: Data-Free Model Compression

    Mohammad Rastegari, AI/ML, Apple

  • 9:45 am – 10:00 am

    Today's Breakout Pitches
  • 10:15 am – 10:45 am

    tiny Talks (consecutive)

    Insights from a multi-purpose self-learning smart sensor
    Speaker: Kaustubh Gandhi, Senior Product Manager Software, Bosch Sensortec

    Presentation title to be announced
    Nitin Chawall, STMicroelectronics

  • 10:45 am – 11:45 am

  • 12:00 pm – 1:00 pm

    Breakout Sessions

    Session #1: Hardware Optimization

    Environmental Noise Classification on microcontrollers
    Jon Nordby, CTO, Soundsensing

    Real-World Performance Analysis of Visual Wake Words
    Luke, Berndt, Senior Director, In-Q-Tel

    Other Breakouts

    • Sponsor Session #2
    • Grants (government agencies & industry)
    • Birds of a feather #2
    • tinyML for Good -- Health (Water, Cancer, Covid, Pandemics)
    Participants to be announced

Friday, March 26

Research Symposium

All times Pacific Standard

Sponsors

2021 tinyML Summit Sponsors

Executive Sponsors

Platinum Sponsors

Gold Sponsors

Silver Sponsors

Sponsorships

Although it will be virtual, the Summit will continue the tradition of high-quality state-of-the-art presentations and we will focus more on the interactivity aspects which we know is important to sponsors. We have been fortunate the last couple of months to be able to attend many virtual events to see what different organizations offer for sponsors, what platforms they use, what works and what doesn’t, etc. In the next few months, we will be researching different platforms to find one that will be most advantageous for both sponsors and attendees. We fill the main benefit to a sponsor of our virtual event will be the number of attendees we expect which easily could be 1,000. And as you can see there are quite a few online/digital benefits for the sponsors which will be woven into the program.

We hope you will join our current sponsoring companies!

tinyML Awards

The tinyML Summit organizers are pleased to announce the creation of three awards to recognize the achievements of the industry and academia toward fulfilling the tinyML vision of ultra-low power machine learning devices at the very edge of the cloud. The awards are:

Best Product of the Year: to be given to the commercial hardware, software, or system product that brought the most significant technological advances in tinyML to the marketplace. Best Product of the Year nomination form

Best Innovation of the Year: to be given to the most innovative concept in either advancing the capability of tinyML devices or leveraging the tinyML approach to create positive societal impact. This award can be given to a product, an invention, or a research result that may lead to potential breakthroughs. Best Innovation of the Year nomination form

Deadline to submit: February 28, 2021

The Best Product of the Year and the Best Innovation of the Year awards are open for nominations between January 15 and February 28, with the winners announced at the tinyML Summit 2021.

Best Paper: to be given to the most outstanding research paper at the tinyML Research Symposium.

No nomination is necessary for best paper as all Symposium papers are automatically considered for the Best Paper award. The innovation described in a Symposium paper (or a previously published paper) can be nominated for the Best Innovation of the Year award.

Breakout Sessions

New for 2021, tinyML is developing a series of breakout sessions. Breakouts are focused on bringing focused topics directly to the audience that needs them.  They will be practical and interactive discussions to foster better understanding of design and application issues, best practices, tools, and funding opportunities to accelerate the deployment of tinyML solutions.  The audience will have the chance to hear from industry leaders in their specific fields and ask questions of them.

Sessions will include:

tinyML Applications

This session will focus on the many diverse market segments where tinyML designs are being deployed such as smart devices to wildlife monitoring.  Real world cases will be described where machine learning meets the rugged demands of ultra-low power, performance and code efficiency.  Industry leaders and innovators will describe how tinyML solved specific problems and review techniques you could use in your designs today.

tinyML Tools Demos

This session will describe a selection of tools available to get you started in hardware and software development utilizing tinyML.  Industry leaders will discuss technology trends, best design practices and what tools are more suited to specific tasks.

tinyML Grants

The explosion of tinyML designs across multiple worldwide applications has opened up opportunities for academia and commercial organizations to seek grant opportunities in many fields.  In this session we will hear from a leading US Government agency on what grant opportunity trends they see in the coming months and years and how to apply.  We will also hear from a grant recipient on how they navigated the grant application process and some of the do’s and don’ts in taking advantage of grants for your tinyML research and projects.

tinyML Summit Sponsor Sessions

The Sponsor sessions will be an opportunity to hear from commercial companies in the tinyML ecosystem on market and technology trends they are addressing to enable the exponential growth of tinyML solutions.  These will not be detailed company product or marketing talks but more interesting discussions on what these companies see happening given their particular vantage points.  Expect to hear how problems and gaps are being solved and what still needs to be done and why.

Breaking News On Disruptive Products And Tools

New for 2021!

Energy efficient embedded machine learning (tinyML) is a fast-growing field of AI technologies and applications including algorithms, hardware, and software capable of performing on-device sensor analytics (vision, audio, IMU, biomedical, etc.). tinyML Summit is an annual conference, drawing attendance from global experts in the field from all the leading and upcoming industry/academia.

The tinyML Summit 2021 offers unrivaled visibility in the tinyML community, it not only represents the industry leading companies, but also represents a vibrant ecosystem of users/designers/researchers of ML. The aim of our community is to facilitate the exchange of ideas, to learn from each other and to help realize and accelerate significant opportunities of tinyML/tinyAI.

This year’s Summit, to be held on March 22-26, 2021, will include an interactive online LIVE experience. In addition to the content above, we are planning an exciting new session, titled "Breaking News On Disruptive Products and Tools". This session allows a small number of time slots (timing TBD) to companies/experts/academia to share the very latest substantial and disruptive developments and upcoming products of significance in the field of tinyML. This can be related to tools, HW IP, algorithms, products, applications, etc.

In addition to providing a platform for presenters to share their disruptive product/algorithm etc. with the audience. We would like to encourage the sharing of demo’s and open source to truly engage the tinyML community.

Given the fast dynamics of this field, the goal here it to select truly the “latest and greatest” developments in tinyML (preferably since November 2020) and offer companies and the academia an opportunity to present these “freshly baked” disruptive products and results to the Global tinyML community. Targeting this goal, only a very limited number of best “disruptive” news items will be accepted. To be considered for inclusion please fill out the form linked below.

Submit Your News Item

Deadline to submit: February 28, 2021

Contact Us

News

TinyML Could Democratize AI Programming for IoT

Upgrading microcontrollers with small, essentially self-contained neural networks enables organizations to deploy efficient AI capabilities for IoT without waiting for specialized AI chips.

read full article

Can artificial intelligence give elephants a winning edge?

Open-source developers and tech giants created the world's most advanced elephant tracking collars.

“Sara Olsson, a Swedish software engineer who has a passion for the natural world created a tinyML and IoT monitoring dashboard”.

read full TechCrunch article

How TinyML Makes Artificial Intelligence Ubiquitous

TinyML is the latest from the world of deep learning and artificial intelligence. It brings the capability to run machine learning models in a ubiquitous microcontroller - the smallest electronic chip present almost everywhere.

read full article

Why tinyML is a giant opportunity

The world is about to get a whole lot smarter. As the new decade begins, we’re hearing predictions on everything from fully remote workforces to quantum computing. However, one emerging trend is scarcely mentioned on tech blogs – one that may be small in form but has the potential to be massive in implication. We’re talking about microcontrollers.

read full description

tinyML book written by Pete Warden and Daniel Situnayake of Google

Neural networks are getting smaller. Much smaller. The OK Google team, for example, has run machine learning models that are just 14 kilobytes in size—small enough to work on the digital signal processor in an Android phone. With this practical book, you’ll learn about TensorFlow Lite for Microcontrollers, a miniscule machine learning library that allows you to run machine learning algorithms on tiny hardware.

read full description

Stanford University Seminar

Evgeni Gousev of Qualcomm and Pete Warden of Google participated in a panel at Stanford University seminar "Current Status of tinyML and the Enormous Opportunities Ahead".

read full article

AI at the Very, Very Edge (EE Times)

When the TinyML group recently convened its inaugural meeting, members had to tackle a number of fundamental questions, starting with: What is TinyML? TinyML is a community of engineers focused on how best to implement machine learning (ML) in ultra-low power systems. The first of their monthly meetings was dedicated to defining the issue.

read full article

TinyML Sees Big Hopes for Small AI (EE Times)

SUNNYVALE, Calif. – A group of nearly 200 engineers and researchers gathered here to discuss forming a community to cultivate deep learning in ultra-low power systems, a field they call TinyML. In presentations and dialogs, they openly struggled to get a handle on a still immature branch of tech’s fastest-moving area in hopes of enabling a new class of systems.

read full article